400k+ bangla news samples, 25+ categories https://www.kaggle.com/furcifer/bangla-newspaper-dataset
Data collected from https://www.prothomalo.com/archive [Copyright owned by the actual source]
The dataset can be used for bangla text classification and generation experiments.
- Starter Bidirectional LSTM (91% test accuracy)
- Weight file https://bit.ly/30qHcxy (Trained with RTX-2080 Ti, tensorflow-gpu, keras)
- Attention
- BERT
- XLNet
- Starter Generative Model